Early in 2006 when my practice was in the process of converting from paper to its first Electronic Health Record (EHR), I distinctly recall the discontent my practice partners and I felt when learning that in this new electronic world we would now be sharing charts with the specialists caring for our patients. Those of us who took great pride in “our” problem lists, were dismayed that the dermatologist or surgeons could now add “their” diagnoses to “our” problem list. We were quickly set straight realizing EHR’s were not “our record,” but the patient’s record. After all, it made total sense for the record to include, when possible, as much information pertaining to the patient, with input from all of the patient’s providers.
Interestingly, 12 years later we are finding the same discomfort among providers who are now being challenged by patient information coming from sources other than the electronic medical record. New data challenges are being exposed as Population Health Management platforms, such as the NextGen Analytics Suite, begin to incorporate multiple new and disparate data sources to create comprehensive, rich, population views and insights.
An example of this can be as basic as data regarding a provider’s attributed patients. Patient attribution was relatively straight forward when the only attribution data to be considered was the EHR data. As we incorporate adjudicated claims data for analysis, this data includes member files which represent the payers view of attributed patients. Providers now find themselves frustrated by patients who “are not my patients” showing up on their quality dashboards. There are multiple reasons for these discrepancies. Sometimes these are simply attribution errors on the part of the payers, though less common, there are still HMO plans which require patients to name a primary care provider even if they have not yet, or ever, seen this provider. Finally, many Medicaid programs will assign patients to primary care providers with the understanding that providers are responsible for the quality and cost of care for these patients even if they have never been seen by the practice or the provider.
Other data challenges stem from data that assigns a diagnosis to a patient which the primary care provider disagrees with. For example, the patient who is assigned a diagnosis of Coronary Artery Disease when they are seen for chest pain in an emergency room, but the primary care provider since has received the results of a normal coronary catheterization test. When claims data is incorporated this patient, based on the claims data, will be added to the Coronary Artery Disease registry.
In spite of these somewhat frustrating data challenges, it is very clear to me that just as in 2006 patients and providers are better served by broader, more inclusive and richer data. That in the case of data, more is truly the merrier. The answer is not to curtail the inclusion of data from disparate sources, rather the onus is on the Population Health Management Analytics platforms to create innovative and actionable tools to reconcile these data challenges. When reconciliation is not possible, the platform should provide ways for these discrepancies to be exposed and viewed. The NextGen Analytics Suite was designed and created to incorporate as many data sources as available. From its inception, it has incorporated multiple attribution algorithms and the ability to create cohorts that reflect discordant attribution methodologies. I look forward to sharing a thorough review of these distinct features and functions.